[FieldTrip] EMG Processing for Coherence Computation

[klocke p.r.o. (pk1e14)]

Dear Jan-Mathjis,
Many thanks for your response.
After a careful walk through the CMC literature, I finds numerous studies that use high-pass filtering of the EMG (before rectification) at 10 Hz and subsequently interpret corticomuscular frequencies below the 10 Hz threshold, e.g. alpha- or theta band coherence.
However, I do not find a methodological discussion on whether the high-pass changes the true coherence, not only by suppression of artefact components, but perhaps also by attenuating (patho-) physiological coherence.
If you happen to have a rationale or a reference in mind justifying this procedure, I would be very grateful to know.



As always, many thanks,

Patrick

On 26. Apr 2024, at 1:13 PM, Schoffelen, J.M. (Jan Mathijs) via fieldtrip <fieldtrip at science.ru.nl<mailto:fieldtrip at science.ru.nl>> wrote:

CAUTION: This e-mail originated outside the University of Southampton.
Hi Patrick,

Please use pubmed to find good practice processing parameters for CMC. Most of the seminal work dates from the end 90’s early 2000’s. You could look up my old work (e.g. Schoffelen et al Science 2005), or look into the references that we used to build upon.

Note that the highpass filter step is applied prior to performing the rectification step. The rectification is needed because we use the EMG as a proxy for the spiking activity of the spinal motor neurons, the firing rhythm of which may be in sync with the EEG/MEG signal obtained from sensorimotor cortical areas. In order to ‘boost’ the information in the EMG signal that pertains to this firing rhythm, one would need to look at the spectral decomposition of the signal’s envelope. The latter is obtained by the rectification. The highpassfilter is needed to remove signal components from the EMG that are non-specific to the firing of the motor neurons. Some people advocate even higher frequencies given that the majority of the energy of EMG spikes lives in a frequency band between 40 and 200 Hz, give or take.

Best wishes,
Jan-Mathijs


On 26 Apr 2024, at 11:01, pk1e14 via fieldtrip <fieldtrip at science.ru.nl<mailto:fieldtrip at science.ru.nl>> wrote:

Dear all,
On the field trip tutorial webpage “Analysis of corticomuscular coherence” it says that EMG data are rectified and high passed at 10 Hz as a standard procedure whereas MEG data is not. Corticomuscular coherence is subsequently computed for frequencies even below 10 Hz.
I was wondering whether there are any previous publications that one can refer to that confirms this procedure?
I know that both the Journal of Journal of Electromyography and Kinesiology stipulate a corner frequency of 10 Hz or the SENIAM recommendations which recommends filters between 10-20 Hz for processing of EMG data but does the same apply to EMG data when calculating coherence?

Many thanks and best

Patrick
_______________________________________________
fieldtrip mailing list
https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
https://doi.org/10.1371/journal.pcbi.1002202

_______________________________________________
fieldtrip mailing list
https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
https://doi.org/10.1371/journal.pcbi.1002202

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20240430/6af63bc8/attachment.htm>